DIY ideas for the 3d printer build

TL;Dr version under 2nd paragraph.



Hi guys! I am new here and have been in the printing scene for about a month now. ive been using the robo r1 and well its had a few kinks and bugs here and there but somehow I managed to get really good at using it that now I can leave it printing for hours and come back to a finished product. this led me to start thinking that I should now build my own printer so that I know the ins and outs of it as well as being able to fully customize it. So my question is, how do I get the printer bed to work like the makergear m2 bed. where the Z and Y axis moves the bed while the x axis moves only the extruder. I would also like to know if you guys think I should go for the Rambo or an arduino mega with ramps 1.4. I am not savvy when it comes down to these types of components but will be learning a lot by working with them and in college when I go in next semester.

ill be using nema 17 stepps and will be trying to get at least 16x16x12 inches of build space on all axes or if I can push it a bit further id like to get about 19x19x16 or something around there. is it even possible? My budget for this printer is a bit limited to about 300-600 per printer. this is incluing the arduino/Rambo board, e3dv6 and other electrical components. the stuff that holds it together I can make from my R1. I also know a machinist who can help me on the base and sturdiness of the build etc.


TL;DR version:
Hi, looking to make printer where the y and z axis moves the print bed whilst the extruder moves only on the x axis. I have an idea of how to set up the rods in the 16x16x12/19x19x16 build which im trying to achieve. its your advice which is what I really want.
could I use arduino mega with ramps 1.4 to do that?
what issues need to be addressed? and what boards do you guys recommend other than what I mentioned that works just as good in your experience for my budget of 300-600

thank you for your time to help me out. I am super newb but am a hands on learner

Generally speaking, you select motors based upon the loads they will see. Big builds, especially with the bed moving in 1 or 2 axes, will require big motors and probably bigger drivers than those that come on the typical printer electronics packages.

If you want a big build envelope avoid moving the bed in any but the Z axis. Cantilevering anything isn't a good idea for a large build envelope printer (and probably not good for a small envelope printer, either).

Take a look at this: [plus.google.com]

Thanks for the reply. I will not go that route then. I will instead do try the z axis only movement, although isn't that cantilevering? unless I place another motor on the other side of the bed, giving me twice the power. so id be using 1 motor for x axis 1 for y and 2 for z. that's viable no? the 5th one being for the extruder

It isn't cantilevered if you lower the Z axis using multiple screws (three points define a plane, so 3 should be enough). That's too much trouble and expense for most builders, so they use a single screw and cantilever the bed. If you have to cantilever something, the slow, small movements of the Z axis are the least problematic. The larger the bed, the more trouble cantilevering will be.

Using multiple motors to move the Z axis leads to problems keeping the bed level. If any of the Z axis motors fails to turn in sync with any other, for any reason, your print becomes trash. The solution is to drive all Z axis screws with a single motor and a continuous loop belt. That's too much trouble and expense for most builders, so they use multiple motors. A quick perusal of this forum will demonstrate the problems with keeping two Z axis motors/screws in sync. Three motors/screws would be a nightmare.

Quote
the_digital_dentist
Using multiple motors to move the Z axis leads to problems keeping the bed level. If any of the Z axis motors fails to turn in sync with any other, for any reason, your print becomes trash. The solution is to drive all Z axis screws with a single motor and a continuous loop belt. That's too much trouble and expense for most builders, so they use multiple motors. A quick perusal of this forum will demonstrate the problems with keeping two Z axis motors/screws in sync. Three motors/screws would be a nightmare.

Unless you drive each of the 3 motors from its own stepper driver. Then you could use a Z probe perform 3-point bed levelling - real bed levelling that is, not bed compensation.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Fixing things for the first layer doesn't guarantee they'll keep turning in sync for every layer of a print- if it did all these people wouldn't have so much trouble keeping the screws working in sync. A small misstep in one screw will quickly ruin the print, and Murphy's law guarantees it will happen about 12 hours into a 13 hour print.

Now if you went to the trouble of putting servos on all the screws, you could be pretty certain that they'd all turn as needed, or at least sound an alarm when they failed.

I think it's easier and more reliable to just belt all the screws together.

i see. great info.

I plan on doing x y movement on the extruder and z axis on the bed. I plan on using 2 lead screws for the bed, so if I use the timing belt to turn both of them. im just wondering if the motor would heat up with the stress. im using the nema17 motor.

also the PSU is something im looking into right now. do you guys have any suggestions on those? what do you think about this one. 12v 30a Power Supply

You have to calculate the required torque for the motor, then get a motor that can deliver that torque. NEMA-17 motors are limited to about 100 oz-in. You can calculate the torque using the motor sizing tools at the Oriental Motor web site.

well im not sure ill be placing 100oz but you never know. ill stick with the nema 17 as I do not have a clue how the azteeg board will deal with powering 3-4 steppers at higher voltages. if I knew more about electronics it would help. what do you guys suggest I use as a psu for those motors? will the PSU linked up above do the job? any recommendation will help. thanks

i might also use a linear actuator as my z axis but don't know how the motor will handle it. so im debating on that. if not i will just buy the acme rods in my cart. theyre pretty pricey at 500mm a rod though

Edited 1 time(s). Last edit at 05/20/2015 01:10AM by davedavedave.

Quote
the_digital_dentist
Fixing things for the first layer doesn't guarantee they'll keep turning in sync for every layer of a print- if it did all these people wouldn't have so much trouble keeping the screws working in sync. A small misstep in one screw will quickly ruin the print, and Murphy's law guarantees it will happen about 12 hours into a 13 hour print.

I disagree. If your XYZ stepper motors skip steps during a print, you've got problems anyway. A skipped step on the X or Y axis corresponds to typically 0.2mm (80 steps/mm), which is highly visible. So I would know if my X or Y motors were randomly skipping steps. The Z motor(s) step much less than the X and Y motors, so they are even less likely to skip steps. There is no reason why multiple Z motors should not stay in sync as long as power is not lost. When power is lost, that's a different matter.

Even if one of my hypothetical multiple Z motors did manage to miss a step, that would only affect the height at that point by 0.004mm (@ 4000 microsteps/mm), which wouldn't make any difference unless it happened many times.

OTOH the typical practice of connecting two Z stepper motors in parallel to one driver output on a RAMPS board usually means that the motors are getting too little current and/or the driver is overheating, which may well lead to missed steps.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

If a motor skips one step, it's going to skip many, and yes, you'll have problems. That's why it is better to keep the screws synced mechanically. There is less to go wrong. If the motor skips, all screws remain synced. You still have to find the problem and fix it, but you won't have to realign the axes.

With multiple screws lifting a bed, especially if there's a screw at each corner, missed steps in one motor result in forces that cause the undercarriage or bed to flex, or to try to flex. That increases the load on the other screws and increases the likelihood of the whole mechanism binding. Even if it doesn't bind, the print quality will surely suffer because the bed will tilt. And what will happen during the next print? Z axis problems are extremely common- just look at the many posts about wobbly prints, and prints not sticking to the bed, etc. in these forums.

With two screws/motors lifting the X axis, if one side missteps the axis alignment will change. It may or may not bind depending on how far off it gets and how sloppy the bearings are and how flexy the guide rails are. There is absolutely nothing good to say about axes that can change alignment so easily. CAD, slicing software, and auto-tramming all assume that the printer's axes are orthogonal and stay that way. Even orthogonal axis compensation assumes that the axes are at least fixed relative to each other. There is NO case in which allowing the axes to change alignment during a print (or otherwise) results in a better print, or increases the reliability of the printing process. Likewise, building in the ability to throw off the axis alignment by easily twisting a screw with your fingers (or your kid's fingers), is bad design/practice.

Either problem is prevented by keeping the screws mechanically synced with gears (generally impractical) or a belt (not too hard or expensive). Before a recent redesign/rebuild, my belt-synchronized printer operated for 2 1/2 years without touching the X axis alignment, I haven't had to touch the alignment since the rebuild was done a couple months ago and I don't expect to have to touch it again before the belt fails, probably many years from now, long after I've given up on FDM printing and moved on to something else. The only way for it to fail is for the belt to break (very unlikely), for a screw on a pulley or the motor mount to come loose (proper assembly prevents this), or for the belt to come off the pulleys (proper assembly prevent this, too), all easily manageable with a wrench during the initial set-up.

There are a LOT of little compromises in 3D printers (usually to drive cost down), each of which contributes a little to unreliability and/or reduction in print quality. When you add all the compromises together, you get the current state of 3D printing for most users- hours of tweaking required to get a decent print, and a generally unreliable process because it depends on too many variables to work just right. The compromises that create the most problems can be avoided once you know what they are. Things like flexy fames/guide rails and multiple motors driving screws that should remain in sync at all times are easily avoided compromises that have a major effect on reliability and quality of 3D printing.

Edited 1 time(s). Last edit at 05/20/2015 09:57AM by the_digital_dentist.

thankjs guys for the info. ive been reading about all the stuff you guys mentioned and been looking heavily into the electronics of 3d printing. ive learned a ton. im still debating on whether or not I should use 2 threaded rods for the z axis or a linear actuator I will make myself. the problem is the stress on the motor. I can add a second one though if needed. but the question would be on what the difference between a threaded rod driven axis vs a timing belt driven axis would be. thinking about it now it seems like it wouldn't be as effective as a threaded rod due to varying tension but its worth a try right? if you guys have any thoughts youd like to share on my z axis plight, youre more than welcome to share.

thanks, Dave

If you're talking about using belts to lift the bed in the Z axis instead of screws, it might be workable. If you're talking about lifting the X axis, I say don't do it. The problem with lifting the X axis becomes apparent when power fails. The axis will come crashing down, smashing the extruder into the print/bed. You could use an electronic clutch to prevent that, and counterweight the X axis so there won't be so much momentum, but that's all getting a bit more complicated than just using a couple screws. On the other hand, if you have a belt drive mechanism from an industrial machine sitting around doing nothing, try it.

Another problem with using belts to do the lifting is the resolution - i.e. the layer thickness. With a 1 mm pitch screw, a ful rev of the drive motor will move the bed 1mm. With a small pulley, one rev will move the bed maybe 25-30 mm. You will need to make a reducer to get reasonable resolution and to amplify the torque from the motor..

Quote
the_digital_dentist
Another problem with using belts to do the lifting is the resolution - i.e. the layer thickness. With a 1 mm pitch screw, a ful rev of the drive motor will move the bed 1mm. With a small pulley, one rev will move the bed maybe 25-30 mm. You will need to make a reducer to get reasonable resolution and to amplify the torque from the motor..

Geared NEMA 17 motors are readily available. They solve both the torque and resolution issues. They're used in the Slid3r, which has an interesting solution for raising the bed using the line/spindle approach, but which could readily be adapted to using a belt.

---

[3DKarma.com] - suppliers of quality, affordable 3D printer kits and filament for the UK market.

seems more complicated than I can handle at the moment. im working on a lot of things and cant focus on the the z axis so I will stick with threaded rods which are plentiful at a store by my place. thanks for the info, ill be sure to keep it in mmind fo rmy next printer when I have more time to fiddle with software and equipment